fix image build bug 1781 (#2200)

Signed-off-by: Yongbozzz <[email protected]>

Author: Yongbozzz

EdgeCraftRAG/Dockerfile.server

@@ -3,7 +3,6 @@ FROM python:3.11-slim
 SHELL ["/bin/bash", "-o", "pipefail", "-c"]
 
 RUN apt-get update -y && apt-get install -y --no-install-recommends --fix-missing \
-    libgl1-mesa-glx \
     libjemalloc-dev \
     libmagic1 \
     libglib2.0-0 \
@@ -16,7 +15,8 @@ RUN wget -qO - https://repositories.intel.com/gpu/intel-graphics.key | \
 RUN echo "deb [arch=amd64,i386 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu jammy client" | \
     tee /etc/apt/sources.list.d/intel-gpu-jammy.list
 RUN apt-get update && apt-get install -y \
-    intel-opencl-icd intel-level-zero-gpu level-zero intel-level-zero-gpu-raytracing \
+    intel-opencl-icd intel-level-zero-gpu \
+    intel-level-zero-gpu-raytracing \
     intel-media-va-driver-non-free libmfx1 libmfxgen1 libvpl2 \
     libegl-mesa0 libegl1-mesa libegl1-mesa-dev libgbm1 libgl1-mesa-dev libgl1-mesa-dri \
     libglapi-mesa libgles2-mesa-dev libglx-mesa0 libigdgmm12 libxatracker2 mesa-va-drivers \

EdgeCraftRAG/tools/quick_start.sh

@@ -51,15 +51,20 @@ function start_vllm_services() {
         pip install -U huggingface_hub
         huggingface-cli download $LLM_MODEL --local-dir "${MODEL_PATH}/${LLM_MODEL}"
     fi
+    HF_CACHE="${HOME}/.cache"
+    if [ ! -d "${HF_CACHE}" ]; then
+        mkdir -p "${HF_CACHE}"
+        echo "Created directory: ${HF_CACHE}"
+    fi
     echo "give permission to related path..."
     sudo chown 1000:1000 ${MODEL_PATH} ${DOC_PATH} ${TMPFILE_PATH}
-    sudo chown -R 1000:1000 ${HOME}/.cache/huggingface
+    sudo chown -R 1000:1000 ${HF_CACHE}
     HF_ENDPOINT=https://hf-mirror.com
     # vllm ENV
     export NGINX_PORT=8086
     export vLLM_ENDPOINT="http://${HOST_IP}:${NGINX_PORT}"
     TENSOR_PARALLEL_SIZE=$(get_user_input "your tp size" 1)
-    SELECTED_XPU_0=$(get_user_input "selected GPU " "0")
+    read -p "selected GPU [$(seq -s, 0 $((TENSOR_PARALLEL_SIZE - 1)))] " SELECTED_XPU_0; SELECTED_XPU_0=${SELECTED_XPU_0:-$(seq -s, 0 $((TENSOR_PARALLEL_SIZE - 1)))}
     DP_NUM=$(get_user_input "DP number(how many containers to run vLLM)" 1)
     for (( x=0; x<DP_NUM; x++ ))
     do
@@ -138,9 +143,14 @@ function start_services() {
             exit 0
         fi
     fi
+    HF_CACHE="${HOME}/.cache"
+    if [ ! -d "${HF_CACHE}" ]; then
+        mkdir -p "${HF_CACHE}"
+        echo "Created directory: ${HF_CACHE}"
+    fi
     echo "give permission to related path..."
     sudo chown 1000:1000 ${MODEL_PATH} ${DOC_PATH} ${TMPFILE_PATH}
-    sudo chown -R 1000:1000 ${HOME}/.cache/huggingface
+    sudo chown -R 1000:1000 ${HF_CACHE}
     HF_ENDPOINT=https://hf-mirror.com
 
     # export ENV